Hugging Face's logo

fastino
/
GLiNER2.5-Decide

Token Classification
GLiNER2
Safetensors
English
extractor
Text classification
Intent classification
Sentiment Analysis
Topic classification
Named Entity Recognition
Model card Files
xet
 Community
9
urchade commited on
Commit
91c8b6b
·
verified ·
1 Parent(s): e4ab809

Upload large classification finetune at step 95500

Browse files
Files changed (7) hide show
  1. README.md +64 -0
  2. config.json +24 -0
  3. encoder_config/config.json +34 -0
  4. model.safetensors +3 -0
  5. special_tokens_map.json +123 -0
  6. tokenizer.json +0 -0
  7. tokenizer_config.json +156 -0
README.md ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: gliner2
3
+ license: apache-2.0
4
+ language:
5
+ - en
6
+ pipeline_tag: token-classification
7
+ tags:
8
+ - gliner2
9
+ - Text classification
10
+ - Intent classification
11
+ - Sentiment Analysis
12
+ - Topic classification
13
+ - Named Entity Recognition
14
+ base_model: fastino/gliner2-large-v1
15
+ ---
16
+
17
+ # GLissify latest
18
+
19
+ Classification-finetuned [GLiNER2 large](https://huggingface.co/fastino/gliner2-large-v1) (340M, DeBERTa-v3-large). Load it with `AutoExtractor`.
20
+
21
+ This checkpoint continues from `fastino/gliner2-large-v1` and was trained on ~2.04M FELIX classification examples. Training used `max_len=2048`, bf16, and an effective batch size of 8. This upload is **step 95500** from the 4x A100 run.
22
+
23
+ ## Install
24
+
25
+ ```bash
26
+ pip install gliner2
27
+ ```
28
+
29
+ ## Classify
30
+
31
+ ```python
32
+ from gliner2 import AutoExtractor
33
+
34
+ model = AutoExtractor.from_pretrained("fastino/GLissify-latest")
35
+
36
+ print(model.classify_text(
37
+ "I need to cancel my flight to Paris.",
38
+ {"intent": ["book", "cancel", "change", "status"]},
39
+ ))
40
+ ```
41
+
42
+ ## Details
43
+
44
+ - **Architecture:** GLiNER2 span extractor (`SpanExtractor`)
45
+ - **Encoder:** `microsoft/deberta-v3-large`
46
+ - **Parameters:** 340M
47
+ - **Base checkpoint:** `fastino/gliner2-large-v1`
48
+ - **License:** Apache 2.0
49
+
50
+ The NER / JSON / relation heads from the base model are still present, but this run optimized classification.
51
+
52
+ ## Citation
53
+
54
+ ```bibtex
55
+ @misc{zaratiana2025gliner2efficientmultitaskinformation,
56
+ title={GLiNER2: An Efficient Multi-Task Information Extraction System with Schema-Driven Interface},
57
+ author={Urchade Zaratiana and Gil Pasternak and Oliver Boyd and George Hurn-Maloney and Ash Lewis},
58
+ year={2025},
59
+ eprint={2507.18546},
60
+ archivePrefix={arXiv},
61
+ primaryClass={cs.CL},
62
+ url={https://arxiv.org/abs/2507.18546},
63
+ }
64
+ ```
config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "_name_or_path": "/home/urchadezaratiana/checkpoints/checkpoint-45500",
4
+ "architecture": "span",
5
+ "architecture_version": 1,
6
+ "architectures": [
7
+ "SpanExtractor"
8
+ ],
9
+ "attn_implementation": "sdpa",
10
+ "config_version": 3,
11
+ "counting_layer": "count_lstm",
12
+ "max_len": null,
13
+ "max_width": 8,
14
+ "model_name": "microsoft/deberta-v3-large",
15
+ "model_type": "extractor",
16
+ "span_head": {
17
+ "dropout": 0.1,
18
+ "max_width": 8,
19
+ "span_mode": "markerV0"
20
+ },
21
+ "token_pooling": "first",
22
+ "transformers_version": "4.48.1",
23
+ "use_moe": false
24
+ }
encoder_config/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "_name_or_path": "/home/urchadezaratiana/checkpoints/checkpoint-45500/encoder_config/config.json",
4
+ "attention_probs_dropout_prob": 0.1,
5
+ "dtype": "float32",
6
+ "hidden_act": "gelu",
7
+ "hidden_dropout_prob": 0.1,
8
+ "hidden_size": 1024,
9
+ "initializer_range": 0.02,
10
+ "intermediate_size": 4096,
11
+ "layer_norm_eps": 1e-07,
12
+ "legacy": true,
13
+ "max_position_embeddings": 512,
14
+ "max_relative_positions": -1,
15
+ "model_type": "deberta-v2",
16
+ "norm_rel_ebd": "layer_norm",
17
+ "num_attention_heads": 16,
18
+ "num_hidden_layers": 24,
19
+ "pad_token_id": 0,
20
+ "pooler_dropout": 0,
21
+ "pooler_hidden_act": "gelu",
22
+ "pooler_hidden_size": 1024,
23
+ "pos_att_type": [
24
+ "p2c",
25
+ "c2p"
26
+ ],
27
+ "position_biased_input": false,
28
+ "position_buckets": 256,
29
+ "relative_attention": true,
30
+ "share_att_key": true,
31
+ "transformers_version": "4.48.1",
32
+ "type_vocab_size": 0,
33
+ "vocab_size": 128011
34
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87f6730124509feafb867898ee88a0c82c236ecb6220276d186610ad2dbf96d6
3
+ size 1945828140
special_tokens_map.json ADDED
@@ -0,0 +1,123 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ {
4
+ "content": "[SEP_STRUCT]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ {
11
+ "content": "[SEP_TEXT]",
12
+ "lstrip": false,
13
+ "normalized": false,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ },
17
+ {
18
+ "content": "[P]",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ {
25
+ "content": "[C]",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ },
31
+ {
32
+ "content": "[E]",
33
+ "lstrip": false,
34
+ "normalized": false,
35
+ "rstrip": false,
36
+ "single_word": false
37
+ },
38
+ {
39
+ "content": "[R]",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false
44
+ },
45
+ {
46
+ "content": "[L]",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false
51
+ },
52
+ {
53
+ "content": "[EXAMPLE]",
54
+ "lstrip": false,
55
+ "normalized": false,
56
+ "rstrip": false,
57
+ "single_word": false
58
+ },
59
+ {
60
+ "content": "[OUTPUT]",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false
65
+ },
66
+ {
67
+ "content": "[DESCRIPTION]",
68
+ "lstrip": false,
69
+ "normalized": false,
70
+ "rstrip": false,
71
+ "single_word": false
72
+ }
73
+ ],
74
+ "bos_token": {
75
+ "content": "[CLS]",
76
+ "lstrip": false,
77
+ "normalized": false,
78
+ "rstrip": false,
79
+ "single_word": false
80
+ },
81
+ "cls_token": {
82
+ "content": "[CLS]",
83
+ "lstrip": false,
84
+ "normalized": false,
85
+ "rstrip": false,
86
+ "single_word": false
87
+ },
88
+ "eos_token": {
89
+ "content": "[SEP]",
90
+ "lstrip": false,
91
+ "normalized": false,
92
+ "rstrip": false,
93
+ "single_word": false
94
+ },
95
+ "mask_token": {
96
+ "content": "[MASK]",
97
+ "lstrip": false,
98
+ "normalized": false,
99
+ "rstrip": false,
100
+ "single_word": false
101
+ },
102
+ "pad_token": {
103
+ "content": "[PAD]",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false
108
+ },
109
+ "sep_token": {
110
+ "content": "[SEP]",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false
115
+ },
116
+ "unk_token": {
117
+ "content": "[UNK]",
118
+ "lstrip": false,
119
+ "normalized": true,
120
+ "rstrip": false,
121
+ "single_word": false
122
+ }
123
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,156 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "[PAD]",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "[CLS]",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "[SEP]",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "[UNK]",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "128000": {
37
+ "content": "[MASK]",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ },
44
+ "128001": {
45
+ "content": "[SEP_STRUCT]",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": true
51
+ },
52
+ "128002": {
53
+ "content": "[SEP_TEXT]",
54
+ "lstrip": false,
55
+ "normalized": false,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": true
59
+ },
60
+ "128003": {
61
+ "content": "[P]",
62
+ "lstrip": false,
63
+ "normalized": false,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": true
67
+ },
68
+ "128004": {
69
+ "content": "[C]",
70
+ "lstrip": false,
71
+ "normalized": false,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": true
75
+ },
76
+ "128005": {
77
+ "content": "[E]",
78
+ "lstrip": false,
79
+ "normalized": false,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": true
83
+ },
84
+ "128006": {
85
+ "content": "[R]",
86
+ "lstrip": false,
87
+ "normalized": false,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": true
91
+ },
92
+ "128007": {
93
+ "content": "[L]",
94
+ "lstrip": false,
95
+ "normalized": false,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": true
99
+ },
100
+ "128008": {
101
+ "content": "[EXAMPLE]",
102
+ "lstrip": false,
103
+ "normalized": false,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": true
107
+ },
108
+ "128009": {
109
+ "content": "[OUTPUT]",
110
+ "lstrip": false,
111
+ "normalized": false,
112
+ "rstrip": false,
113
+ "single_word": false,
114
+ "special": true
115
+ },
116
+ "128010": {
117
+ "content": "[DESCRIPTION]",
118
+ "lstrip": false,
119
+ "normalized": false,
120
+ "rstrip": false,
121
+ "single_word": false,
122
+ "special": true
123
+ }
124
+ },
125
+ "additional_special_tokens": [
126
+ "[SEP_STRUCT]",
127
+ "[SEP_TEXT]",
128
+ "[P]",
129
+ "[C]",
130
+ "[E]",
131
+ "[R]",
132
+ "[L]",
133
+ "[EXAMPLE]",
134
+ "[OUTPUT]",
135
+ "[DESCRIPTION]"
136
+ ],
137
+ "backend": "tokenizers",
138
+ "bos_token": "[CLS]",
139
+ "clean_up_tokenization_spaces": false,
140
+ "cls_token": "[CLS]",
141
+ "do_lower_case": false,
142
+ "eos_token": "[SEP]",
143
+ "extra_special_tokens": {},
144
+ "is_local": false,
145
+ "local_files_only": false,
146
+ "mask_token": "[MASK]",
147
+ "model_max_length": 1000000000000000019884624838656,
148
+ "pad_token": "[PAD]",
149
+ "sep_token": "[SEP]",
150
+ "sp_model_kwargs": {},
151
+ "split_by_punct": false,
152
+ "tokenizer_class": "DebertaV2Tokenizer",
153
+ "unk_id": 3,
154
+ "unk_token": "[UNK]",
155
+ "vocab_type": "spm"
156
+ }